[burger]

Another non peer reviewed study. We’ve had way too many of those. With all sorts of conclusions.  So many unknown unknowns.

I only skimmed the study but it looks careful with a relatively sophisticated and appropriate analysis.  That's not to say that these results will replicate in other study populations, but they split their sample into testing and training pools and got fairly good replication of the results.  That's about as good as you can get for this kind of research.  And it could provide an interesting way to assess patients who might be headed for bad outcomes for early intervention.  That could, if this holds up, save lives.

More generally, disbelieving everything not peer reviewed is not a tenable approach to assessing covid 19 research. There are some bad preprints for sure. But a few of the peer reviewed covid studies are also crap, so even peer review does not guarantee good results.  Everything should be evaluated on its own merits.

[burger]

Another non peer reviewed study. We’ve had way too many of those. With all sorts of conclusions.  So many unknown unknowns.

I only skimmed the study but it looks careful with a relatively sophisticated and appropriate analysis.  That's not to say that these results will replicate in other study populations, but they split their sample into testing and training pools and got fairly good replication of the results.  That's about as good as you can get for this kind of research.  And it could provide an interesting way to assess patients who might be headed for bad outcomes for early intervention.  That could, if this holds up, save lives.

More generally, disbelieving everything not peer reviewed is not a tenable approach to assessing covid 19 research. There are some bad preprints for sure. But a few of the peer reviewed covid studies are also crap, so even peer review does not guarantee good results.  Everything should be evaluated on its own merits.